As of November 2021, the APT Forum is closed to new posts. Like with many online forums, usage has decreased in recent years. All previous posts are still available.

Steamin' Steve gets STACKED!

edited April 2018 in Specific Hand Questions

Hey @SteveStokely,
Was this a "pot committed" puke call? :) Hero has 4s. You don't. Hero has all the sets.You can have QQ. 56s gets there. Hero knows you have top two with at least AQ so not shoving 2 pr. (Hero never has AA or Kk with no 3bet.)

What? Can't fold a big pair? Haha

All comments welcome from both players' perspectives.



  • 1warlock

    After 2 check-calls to rather sizeable bets, I think the bot should check back the river with the overpair. From your spot, I may have looked to get in a raise on the turn to extract value from overpairs and AQ but letting the bot barrel is fine as well. I'm not sure about the check-shove on the river because I wouldn't think 1 pair hands could call but the bot did so what do I know?

    As a point of interest, I find the advice across all streets to be pretty inexplicable, especially the advice to fold pre-flop and then again on the river. Way too tight, IMO.

  • highfive

    Thanks @1warlock. Insightful post as usual.
    Surprisingly 54s has 31.4 equity against a tight utg range of AQ+ 10s+ and i'm not sure 10s are in there.
    This is an obvious implied odds play. I expect 10/1 at least so easy call heads up. Fully expect to check fold mostly on flop.
    I'm always told in the Training Plan that I defend too much. BUT i'm very profitable from the blinds. I'm in the +30-40 bb/100 area while most players on APT are -bb/100. <--- their stats not my guess. So of course I trust my blind strat over the advice.
    On to the hand. I'm never folding here. I shove expecting to win against all holdings except QQ.
    KGB bots never have 56, 33, 77, or any 4s here.
    Never say never in poker but KGB bots are super nits utg. In this case I 'm quite sure of their range.
    Big win because KGB is so tight. It's quite difficult to extract chips.

    Just posted the hand to give @stevestokely the business.

  • 1warlock
    edited April 2018

    No worries. I don't know this bot's opening range UTG but a solid HU value calling range from the BB vs a 13% EP open looks like this: 99-22, ATs-A7s, K9s+, Q9s+, J9s+, T8s+, 98s, 87s, 76s, 65s, 54s, AJo-A9o, KTo+, QTo+, JTo. Add in the 3-bets for TT+, AJs+ and AQo+ and you are defending top ~27% of hands. 4/5s is within this range, as noted. If the bots are tighter than this, then 4/5s would likely fall out of the range but not by much. Even if it did, if you have superior post-flop abilities, you can add back hands and defend wider.

    When I joined here I asked about the bots and was told very frankly that they do not play anything close to optimally. They do however represent a good cross section of non-professional players. So while I don't love its bet on the river, its not unrealistic to expect some human players to make the very same play.

    I was thinking about the advice generated. I wonder if it becomes less accurate as the hand progresses once the human player deviates from a suggestion? Does it re-evaluate each street, adjusting for the deviations? If not, is it only evaluating relative hand strengths at each decision point? I just don't know how it could come up with a fold on that river with Hero behind to only 6 of 180 possible combinations (77, QQ).

    Added: I would love to be able to run hands mirror-image and see what the computer generated advice would have told the bot to do facing the check-raise on the river. If it advised you to fold trips facing the bet, I doubt it would have told the other player to call the shove with just the over-pair.

    Also, another player brought up the fact that the bots play too tight/passive in the standard mode. Have you changed the settings to have them play more aggressively than normal? This seems to help replicate live players a bit more realistically. I assume it also increases their opening ranges.

Sign In to comment.